Neonatal hearing screening using otoacoustic emissions elicited by maximum length sequences

The aim of this study was to provide statistical information on otoacoustic emission (OAE) measurements in new-born infants, which could be useful in the interpretation of results and could add some clinical parameters that might be used for future objective and automatic response evaluations. Transient evoked OAEs (TEOAEs) using maximum length sequences (MLS) were recorded in 129 full-term new-born infants between 48 and 72 hours of age. The use of MLS enabled stimulus rates of up to 2000/sec to be utilized. The analysis comprised five main issues: (1) signal to noise ratio (S/N); (2) derived non-linear responses; (3) a combination of derived non-linear emissions called compound non-linear emissions; (4) time/frequency analysis and (5) signal power analysis. The result of the analysis were as follows. (1) The largest S/N ratio was obtained when more than 60% of the recordings were rejected. (2) The derived non-linear emissions comprised level non-linear and rate non-linear responses, the latter obtained by subtraction of recordings at fixed stimulus level, but at different stimulus rates. The rapid suppression of emission amplitude as the stimulus rate increases is the basis for this non-linear response which also features stimulus cancellation. (3) The compound non-linear emissions, consisting of a sum of derived non-linear responses may prove efficient by proper selection of stimulus parameters, but did not in the present form show the expected improvements of amplitude and stimulus cancellation. (4) The time/frequency analysis, however, based on calculations of the instantaneous frequency as a function time after stimulus onset, showed a consistent tendency of falling frequency versus time after stimulus onset as a basic characteristic feature of OAE. (5) The analysis of the emission power as a function of time after stimulus onset showed that linear recordings in new-born infants are reliable, with stimulus ringing fully decayed 3 ms after the stimulus onset. The findings may prove useful in future attempts to develop screening equipment with automatic response evaluation for new-born infants. From a clinical point of view the MLS technique was successful when measuring emissions in neonates, owing to the fast recording time and the utilization of an efficient noise rejection paradigm.